Modeling Rainfall and Groundwater level data Using Time-varying copula models
Non-stationary data are often created when the observations of a study are collected sequentially in a time-dependent structure. In such a case, there will usually be a time trend with abrupt changes in the average or/and variance of the observations, which indicates that the data is non-stationary....
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | fas |
| Published: |
University of Isfahan
2024-07-01
|
| Series: | ریاضی و جامعه |
| Subjects: | |
| Online Access: | https://math-sci.ui.ac.ir/article_28524_67028d7f16b54131ed7e3b0d83e345c7.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Non-stationary data are often created when the observations of a study are collected sequentially in a time-dependent structure. In such a case, there will usually be a time trend with abrupt changes in the average or/and variance of the observations, which indicates that the data is non-stationary. To describe such data using statistical distributions and fitting parameters, time-varying models are suitable. The aim of this study is to introduce and apply time-varying models in which parameters are considered as time-varying in both marginal distributions and copula models. According to the monthly collection of rainfall and groundwater level data, the nature of these data is time-varying and the trend changes in these data shows that the average of data changes abruptly over time. To describe the correlation structure between these data, marginal distributions and then time-varying copulas have been used, so that the parameter of these models is considered to be vary over time as a function of time in the form of polynomial or exponential regression functions. |
|---|---|
| ISSN: | 2345-6493 2345-6507 |